linux-kernel - Re: [PING^2][PATCH] RISC-V: PCI: Avoid handing out address 0 to devices

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <alpine.DEB.2.21.2206231750490.57474@angie.orcam.me.uk>
Date:   Fri, 24 Jun 2022 18:18:11 +0100 (BST)
From:   "Maciej W. Rozycki" <macro@...am.me.uk>
To:     Palmer Dabbelt <palmer@...belt.com>
cc:     Paul Walmsley <paul.walmsley@...ive.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PING^2][PATCH] RISC-V: PCI: Avoid handing out address 0 to
 devices

On Wed, 22 Jun 2022, Palmer Dabbelt wrote:

> > > Therefore avoid handing out address 0, by bumping the lowest address
> > > available to PCI via PCIBIOS_MIN_IO and PCIBIOS_MIN_MEM up by 4 and 16
> > > respectively, which is the minimum allocation size for I/O and memory
> > > BARs.
> > 
> >  Ping for:
> > <https://lore.kernel.org/lkml/alpine.DEB.2.21.2204271207590.9383@angie.orcam.me.uk/>
> 
> Sorry, I got this mixed up with the non-RISC-V patch.

 If you mean this:

<https://lore.kernel.org/lkml/alpine.DEB.2.21.2202260044180.25061@angie.orcam.me.uk/>

then we just don't have consensus to move forward.  If we ever do for a 
generic change, then we can revert the RISC-V platform solution, as it's 
merely an internal implementation detail and not a part of the ABI or 
something.

>  David poked me about
> it, this is on for-next.  It's passing my tests, but they're just QEMU so
> probably not all that exciting here.

 Thanks!  I don't know offhand what QEMU supports as far as the RISC-V 
architecture is concerned; I guess you can't just enable a PCI port-I/O 
serial port in the simulator and see if it works with Linux or not.

 Anyway it's just number shuffling, so the change should be reasonably 
safe.

  Maciej